English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Recently, pretrained language models (e.g., BERT) have achieved great successon many downstream natural language understanding tasks and exhibit a certainlevel of commonsense reasoning ability. However, their performance oncommonsense tasks is still far from that of humans. As a preliminary attempt,we propose a simple yet effective method to teach pretrained models withcommonsense reasoning by leveraging the structured knowledge in ConceptNet, thelargest commonsense knowledge base (KB). Specifically, the structured knowledgein KB allows us to construct various logical forms, and then generatemultiple-choice questions requiring commonsense logical reasoning. Experimentalresults demonstrate that, when refined on these training examples, thepretrained models consistently improve their performance on tasks that requirecommonsense reasoning, especially in the few-shot learning setting. Besides, wealso perform analysis to understand which logical relations are more relevantto commonsense reasoning.

Teaching Pretrained Models with Commonsense Reasoning: A Preliminary KB-Based Approach